Spatiotemporal characterization of cellular tau pathology in the human locus coeruleus-pericoerulear complex by three-dimensional imaging

Abstract

Tau pathology of the noradrenergic locus coeruleus (LC) is a hallmark of several age-related neurodegenerative disorders, including Alzheimer's disease. However, a comprehensive neuropathological examination of the LC is difficult due to its small size and rod-like shape. To investigate the LC cytoarchitecture and tau cytoskeletal pathology in relation to possible propagation patterns of disease-associated tau in an unprecedented large-scale three-dimensional view, we utilized volume immunostaining and optical clearing technology combined with light sheet fluorescence microscopy. We examined AT8⁺ pathological tau in the LC/pericoerulear region of 20 brains from Braak neurofibrillary tangle (NFT) stage 0-6. We demonstrate an intriguing morphological complexity and heterogeneity of AT8⁺ cellular structures in the LC, representing various intracellular stages of NFT maturation and their diverse transition forms. We describe novel morphologies of neuronal tau pathology such as AT8⁺ cells with fine filamentous somatic protrusions or with disintegrating soma. We show that gradual dendritic atrophy is the first morphological sign of the degeneration of tangle-bearing neurons, even preceding axonal lesions. Interestingly, irrespective of the Braak NFT stage, tau pathology is more advanced in the dorsal LC that preferentially projects to vulnerable forebrain regions in Alzheimer's disease, like the hippocampus or neocortical areas, compared to the ventral LC projecting to the cerebellum and medulla. Moreover, already in the precortical Braak 0 stage, 3D analysis reveals clustering tendency and dendro-dendritic close appositions of AT8⁺ LC neurons, AT8⁺ long axons of NFT-bearing cells that join the ascending dorsal noradrenergic bundle after leaving the LC, as well as AT8⁺ processes of NFT-bearing LC neurons that target the 4th ventricle wall. Our study suggests that the unique cytoarchitecture, comprised of a densely packed and dendritically extensively interconnected neuronal network with long projections, makes the human LC to be an ideal anatomical template for early accumulation and trans-neuronal spreading of hyperphosphorylated tau.

Keywords: Alzheimer’s disease; Locus coeruleus; Tau pathology; Three-dimensional, iDISCO, light sheet fluorescence microscopy.

Fig. 1

3D segmentation and cytoarchitecture of the human LC/PC complex. a–c 500 μm-thick, representative coronal optical slices with colour-coded segmented subregions. Subcoeruleus (subC): green; locus coeruleus core: blue; A4—pars cerebellaris: red; locus coeruleus shell: pink. me5: mesenchepalic trigeminal tract; 4th: fourth ventricle. d–e 3D reconstruction of the same block (sagittal view). Arrow: an artery that enters the LC core. f–i Representative 3D crops (500 × 500 × 500 μm) (f, h) and higher resolution 3D crops (200 × 200 × 200 μm) in blend rendering mode, detailing the morphology of the somato-dendritic compartment of NA neurons (g, i) from the caudal (f–g) and rostral (h–i) ends of a Braak 0 block. j Higher resolution of duo (*) and triplet (**) cells with ‘hugging’ somas from a 20 μm-thick horizontal optical slice. k 3D reconstruction of a cell duo with ‘hugging’ somas (blend rendering mode). l Reconstruction of the complex dendritic network of LC core NA neurons (surface with 2 μm grain size). m Dendro-dendritically closely apposed adjacent TH⁺ neurons (blend 3D rendering mode). Arrows: dendritic appositions between neurons #1– #2 and #3– #4. n–p Quantifications of TH⁺ cells (Braak 0 brains). N = 6; *P < 0.05 (vs. shell), ^#P < 0.05 (vs. A4), ⁺P < 0.05 (vs. subcoeruleus). Data expressed as mean ± SEM. Each micrograph shows TH volume immunostaining from Braak 0 cases. Scale bars are indicated in each micrograph

Fig. 2

Large-scale quantitative analysis of the total AT8⁺ immunostaining volume and AT8⁺ cells. a Analysis pipeline. b–d Summary and graphical representation of quantified AT8⁺ immunostaining. AT8⁺ immunostaining volumes normalized for subregion volumes (b). Number of AT8⁺ neurons normalized for subregion volume (c) and for TH⁺ cell number (d). Two-way ANOVA main effect F, dF and P values are indicated in each panel. N = 6 (Braak 0), N = 6 (Braak 1–2), N = 5 (Braak 3–4), N = 3 (Braak 6); *P < 0.05 (vs. Braak 0), ^#P < 0.05 (vs. Braak 1–2), ⁺P < 0.05 (vs. Braak 3–4). Data expressed as mean ± SEM. e Representative images demonstrating full tau staining volumes (white in the left column) and AT8⁺ cells (as dots in the right column). Subregion volumes are colour-coded as shown in the upper right panel. subC: subcoeruleus. B0, B1-2, B3-4, B6 = Braak NFT stages 0, 1–2, 3–4, 6, respectively. Scale bars are indicated in each micrograph

Fig. 3

Spatial relation of AT8⁺ cell bodies and processes. a Segmentation pipeline of AT8⁺ cell bodies and processes. b–e Proportional distribution of AT8⁺ volumes for cell bodies and processes in LC/PC subregions throughout the Braak stages. N = 6 (Braak 0), N = 6 (Braak 1–2), N = 5 (Braak 3–4), N = 3 (Braak 6). *P < 0.05 (vs. LC core cell body volume %). Data expressed as mean ± SEM. f–g Representative images demonstrating the segmented AT8⁺ cell body volumes (white surface) and process volumes (red surface) in the LC core (transparent blue surface in f) and in the subcoeruleus (transparent green surface in g) from a Braak 3 subject. h Representative image demonstrates in-depth spatial analysis of processes around cell bodies in the LC core in a Braak 0 subject. The 3D space around cell bodies (white surfaces) was divided into concentric, 100 μm-wide spheres, and the AT8⁺ process volume was determined in each sphere. The portions of process volumes in certain spheres are colour-coded as indicated in the lower left corner. i–l Quantification of the spatial distribution of AT8⁺ processes around AT8⁺ cell bodies in the LC core (i), shell (j), A4 (k) and subcoeruleus (l) in Braak 0–2 subjects. N = 12; *P < 0.05 (vs. the 50–150 μm-range values). Data expressed as mean ± SEM. m Illustration of AT8⁺ processes (red) that leave the LC core (blue surface); coronal perspective from caudal view in a Braak 2 subject. B0, B1-2, B3-4, B6 = Braak NFT stages 0, 1–2, 3–4, 6, respectively. Scale bars are indicated in each micrograph

Fig. 4

AT8⁺ cellular structures in 3D and their semiquantitative assessment. a–s Representative images of AT8⁺ structures on the cellular level, identified by volume imaging. The orange arrows indicate a hypothetical direction of neurofibrillary tangle maturation. Cell with low-intensity staining (Braak 2 subject) (a). Cell with intact processes and inhomogeneously filled soma (Braak 0 subject) (b). Cell with intact processes strongly and homogeneously filled soma (Braak 2 subject) (c). ‘Stellate-like’ cell (Braak 2 subject) (d). Cell with partially atrophic dendrites (Braak 2 subject); arrows: atrophic, swollen dendritic branch; arrowheads: morphologically intact axon (e). Cells with severely atrophic processes (Braak 3 subjects) (f–g). Matured tangle-like, flame-shaped cell (*); arrow: swollen axon fragment; (Braak 6 subject) (h). Matured tangle-like cell with flame-like morphology (*), matured tangle-like cells with ball-like morphology (**), (Braak 6 subject) (i). Cell with fine filamentous somatic protrusions (Braak 2 subject) (j). Same cell as in j but shown in 2D projection and higher resolution; arrows: dendrites; arrowheads: fine somatic protrusions (j’). Transition form with fragmenting somatic protrusions (Braak 2 subject) (k). One cell with advanced fragmentation of processes and somatic protrusions (*) and another cell with advanced process atrophy (**), (Braak 3 subject) (l). Disintegrating cell/soma (arrow), (Braak 2 subject) (m). Same cell as in m but shown in 2D projection with higher resolution (TH channel, gray); arrow: disintegrating soma; *adjacent intact TH⁺ cells (n). Same cell as in m and n but shown in 2D projection with higher resolution (AT8 channel, red); arrow: disintegrating soma; arrowheads: fragmenting dendrites (o). Long, morphologically intact axons (Braak 0 subject) (p). Long axon with swollen varicosities (Braak 0 subject); arrowheads: swollen varicosities (q). Swollen axonal fragments (Braak 3 subject) (r). Debris of fragmenting dendrites from a Braak 2 subject; *cell body of AT8⁺ cell; arrowheads: fragmenting dendrites of the same cell (s). (t) Semiquantitative scoring of the cellular AT8⁺ structures. This table is the summary of Supplementary table 3. The dominating cellular forms in certain Braak stages are indicated with red marks. Scale bars are indicated in each micrograph

Fig. 5

More advanced tau pathology in the dorsal vs. ventral part of the LC core. a AT8⁺ immunostaining is more abundant in the dorsal part of the LC core (white surface) than in the ventral part (red surface) throughout the Braak stages. b–c Proportional distribution of AT8⁺ immunostaining volume along the dorso-ventral axis, normalized for segment volume (b) and TH⁺ cell number (c). N = 19; *P < 0.05. Data are represented as individual data points and as mean ± SEM. d More AT8⁺ cells (white dots) are found in the dorsal LC core segment (blue surface), compared to the ventral LC core segment (red surface). e–f Proportional distribution of AT8⁺ cells along the dorso-ventral axis, normalized for segment volume (e) and TH⁺ cell number (f). N = 19; *P < 0.05. Data represented as individual data points and as mean ± SEM. g–i Demonstration of the dorso-ventral distribution of AT8 volume staining in a 1000 μm-thick optical slice of LC core (Braak 6). Boxes in g indicated with h’ and i’ are enlarged in h and i, respectively. (j) Semiquantitative scoring of cellular AT8⁺ structures in the dorsal (upper table) and ventral (lower table) LC core segments. Blue marks: structure is more abundant in the dorsal segment. Red marks: structure is more abundant in the ventral segment. B0, B1-2, B3-4, B6 = Braak NFT stages 0, 1–2, 3–4, 6, respectively. Scale bars are indicated in each micrograph

Fig. 6

In-depth spatial analysis reveals a clustering tendency of AT8⁺ cells in the LC core. a Paired comparisons of Monte-Carlo simulated and actual (real) mean nearest neighbour distances throughout the Braak stages. N= 5 (Braak 0), N  = 6 (Braak 1–2), N = 5 (Braak 3–4), N = 3 (Braak 6); *P < 0.05. b Graphical representation of nearest neighbour index (NNI) calculations. Cases with values below 1.0 exhibit a clustering tendency of AT8⁺ cells. Data are represented as individual data points and as mean ± SEM. c Distribution of dense AT8⁺ cells along the dorso-ventral axis; N = 18; *P < 0.05. Data are represented as individual data points and as mean ± SEM. d–e Demonstration of dense cells (big coloured spots) and non-dense cells (small white dots) in the volume of a Braak 0 (d) and a Braak 3 (e) case. Dense cells were clustered and colour-coded by the ‘split spot’ MATLAB Imaris XTension: cells that belong to the same ‘duo’ or ‘minigroup’ have the same colour. f–h Examples of dense (clustering) cells. AT8 channel, white dots indicate cell bodies (f, g, h). Coloured spots in f’, g’ and h’ refer to the AT8⁺ cells in panels f, g and h, respectively. Distances between the centers of cell bodies (in μm) are indicated in f’, g’ and h’. i Volume rendering of AT8⁺ neurons with an indication of dense cells, representative 3D crop (1.8 × 1 × 1 mm overview, Braak 1 brain). j–o Quantitative characterization of dense cells. N = 5 (Braak 0), N = 6 (Braak 1–2), N = 5 (Braak 3–4), N = 3 (Braak 6); *P < 0.05 (vs. Braak 0); ^#P < 0.05 (vs. Braak 1−2). Data expressed as mean ± SEM. Scale bars are indicated in each micrograph

Fig. 7

Potential routes for spreading of hyperphosphorylated tau: AT8⁺ cell appositions and long axons. a–d Dendrites from pairs of AT8⁺ neurons that are closely apposed in the LC core from two Braak 0 subjects (a–a’ and c), from a Braak 1 subject (b–b’) and from a Braak 6 subject (d). a’ and b’ are 2D-projected representations of the 3D-rendered cells shown in a and b, respectively. Arrowheads in all panels point to the dendritic appositions. (e–e’) 3D rendering of two neighbouring duo cells with apparent ‘tunnel-like’, thin interconnections (arrowheads) (e) and a 2 μm-thick optical slice showing the same cells (e’). (f) AT8⁺ axons arising from the LC core (blue surface) and joining the dorsal NA bundle ventrally from the LC. g–h Traced AT8⁺ axons of LC core neurons. Axons from cell bodies (coloured dots) usually take a U-turn before leaving the LC core, typically in the ventral direction, running towards the dorsal NA bundle (‘red’ cell in g, and ‘red’ and ‘green’ cells in h), or rarely dorsally, towards the A4 (‘yellow’ cell in h). Scale bars are indicated in each micrograph

Fig. 8

AT8⁺ LC dendrites join the subependymal NA plexus below the 4th ventricle. a–b The subependymal NA plexus below the 4th ventricle with terminal dendritic segments (*) and dendrites (arrowheads in b) from LC NA cell bodies (arrows in b), TH channel. c–d AT8⁺ NA neuronal soma (arrow in d) in the LC core extending hyperphosphorylated tau-filled dendrite (arrowheads in d) to the subependymal NA dendritic plexus (Braak 0 subject). Red and whitish grey represent AT8 and TH volume staining, respectively. c: AT8 channel only; d: AT8 + TH channels together. e 2D projection of the AT8⁺ neuron shown in c, AT8 channel. f–h Dense hyperphosphorylated tau-filled subependymal dendrites in 300 μm-thick optical slices, in a Braak 3 (f) and a Braak 6 (g–h) subject. Box indicated with h’ in g is enlarged as h. Hyperphosphorylated tau-filled terminal dendritic segments (arrowheads in h) approach the ventricle surface to 40–50 μm. The dashed line in h indicates the ventricle surface. AT8 channel. Scale bars are indicated in each micrograph